Accurate chromosome segregation depends on bi-orientation (i.e., sister kinetochores attaching to spindle microtubules (MTs) from opposite poles), which relies on tension-dependent stabilization of kinetochore MTs. Multiple lines of evidence suggest that Aurora B kinase, the enzymatic component of the chromosome passenger complex (CPC), is a key element of this mechanism. Strikingly, CPC binding sites are enriched at the inner centromere, not at the outer kinetochore where Aurora B substrates bind MT ends. Moreover, phosphorylation of these kinetochore substrates, which reduces MT binding, decreases as the distance from the centromere to the kinetochore increases with tension. These observations established a correlative link between Aurora B signaling and tension and led to an intuitively attractive spatial separation model, in which distance-dependent phosphorylation plays a crucial role in bi-orientation. However, the mechanisms underlying distance-dependent phosphorylation by Aurora B are unknown. Furthermore, whether tension can be sensed by Aurora B-independent pathways in vivo, as suggested by in vitro experiments has not been tested. Aurora B is known to auto-activate by auto-phosphorylation in trans, so we developed a quantitative model to describe the resulting non-intuitive spatial nonlinear dynamics. Our model predicts that distance- dependent phosphorylation by Aurora B is established by a reaction-diffusion mechanism. The essential features of this mechanism are: 1) Aurora B auto-activates at centromeres due to a high density of CPC binding sites (i.e., clustering), and 2) this activity propagates to kinetochores by unbinding of active kinase, diffusion, and kinase activation/inactivation reactions in solution, which depend on the soluble kinase/phosphatase ratio. Aim 1 will test the hypothesis that Aurora B regulates kinetochore-MT interactions through a reaction-diffusion process in vivo. We will manipulate CPC clustering at centromeres, the kinetics of centromere unbinding, and the soluble kinase/phosphatase ratio, and measure the effects of these perturbations on kinetochore function. Aim 2 will reconstitute Aurora B phosphorylation dynamics and spatially-regulated MT binding in vitro. The in vivo situation is complex, and in vitro reconstitution will allow us to establish direct, quantitative relationships between the inputs and the behavior of the system. This aim builds on an in vitro system that we developed using recombinant Aurora B and phosphatase and fluorescent substrates. Aim 3 will define the contributions of tension and Aurora B to regulating kinetochore MTs in vivo. To uncouple tension from Aurora B, we developed a chemically-induced dimerization strategy that allows us to directly manipulate Aurora B activity at kinetochores and independent of tension. We will (1) determine how controlled changes in Aurora B activity at kinetochores affect MT dynamics, and (2) dissect the direct effects of tension and indirect effects via Aurora B. The result of these experiments will b detailed understanding of how both biochemical and mechanical changes at kinetochores control interactions with MTs.